aerodynamic effects

[MikeSt]

I'd guess most models built by the average person are going to be within maybe 1mm or .5mm, and maybe .5 degrees for angles when it comes to accuracy. Probably more than good enough for a good flying model. (How precisely are full scale planes assembled ?)

I'm curious as how to calculate or estimate the effects of imprecise construction.

Not that I want to be sloppy ;) ... . I'm interested in the aerodynamics of it !

Take some extreme examples.

Wingpan 10". Chord 4".

1) Left wing span is only 9" ...or 8" ... 7 . .6 .. 5.

When does the plane become unflyable ?

2) Back to normal, but left wing chord is now 3.9" .. ... 3.5" .. 2".. 1"

3) Normal but, left wing is further forward/back along the fuselage than the right wing

4) Normal but, left wing has more/less dihedral than the right wing.


5) Same things for the stab. E.g. left stab span is less than right, ... left stab chord less than right


I'd imagine there could be some heavy math involved:D, if you know it and feel inclined, lay it on me ;)

Mike

[steve lewin]

You can make some amazingly asymmetrical designs fly, just ask Burt Rutan. But how you would calculate the effects of all those gross errors on a plane designed to be symmetrical I can't imagine. It doesn't seem like the sort of work anyone would ever have done, it being a little difficult to imagine anyone accidently building one wing half size and then thinking "Hey, I'll fly it anyway".

In many cases you're just describing ways to make one side of the plane produce a different amount of lift from the other. The effects of that are fairly obvious and predictable aren't they ?

Steve

[vintage1]

Full scale pilots often repoet hat older planes fly crabwise, or nose down, or one wing down.

Basically, they are bent. (the planes).

[MikeSt]

Ok, what I was trying to get at was how precise a plane needs to be built.

On a 500mm (20 in ) wingspan 1 or 2 mm probably is not going to make a big difference if at all.

But 5 mm, 10 mm ...? At some point it's gonna make a difference.

[steve lewin]

My guess, and that's all it is, is that you could easily be 5% out on almost anything without any noticeable effect on the flight characteristics. So your 500mm plane could have one wing 12.5mm shorter than the other and you'd never notice (unless it was intended to be a precision pattern ship). Some aspects of a model are a lot less sensitive than that.

It's not that there is no effect on the aerodynamics it's just that most pilots are not very sensitive to such things. For example, having one wing short would probably need one click of trim to keep it level. It may need a tiny bit more aileron to turn one way than to turn the other. You'd never even spot it.

If you look at some free flight planes and the changes you have to make to get their flight patterns right you'll see what I mean. E.g. I was playing with a little indoor rubber model this weekend. It has one wing 8% longer than the other and the "horizontal" stab at about 20 degrees to the horizonal and about 10 degrees offset on the vertical fin. It just circles nicely.

Long before the plane becomes aerodynamically unpleasant you'll have given up because it looks so silly ;).

Steve

[vintage1]

Well, my point was that anything makes a difference, and bigger things make bigger differences.

I think you are looking for a discrete solution to a continuous function...if you wanted to ask a more specific question, like 'how twisted does a wing have to be before top speed is reduced by 3%' then I suspect, for a given plane, there exists a reasonably exact answer.

As an example, if you mount a dihedral wing a bit crooked, (skewed) then your model will fly slightly sideways. And possibly one wing down. The same is not *as* true of a fully symmetrical wing with zero dihedral, which will tend to have a roll, rather than yaw, moment introduced by that defect.

As a youngster I built some appallingly warped models. Gliders - constant speed - could be trimmed reasonably. Rubber models, already suffering from huge torque moments, were generally untrimmable. Power planes flew badly, but had enough power to at least fly..

I guess I try and build to within one or two degrees of 'straight' - which is about 2-4% 'skew' on things like dihedral angle and so on, and better than that on things like wing plans - I mean you ought to be within say 3mm on a meter span, which is .3%, or 0.2 degrees give or take.

Let's face it, thats about as good as you *can* build. Pattern flyers end up with little wedges glued here and there to fine tune their models - but how far do you want to go? What is the weight tolerance on your motors and bits of wood? what wood shrinkage/expansion do you get on a humid/dry day (could be up to 1% you know:) ) - How accurately do your RPMs and therefore torque and propwash stay teh same from one flight to the next, from one fuel tank to teh next?

[Sparky Paul]

Your sample plane is really too small to be a good indicator of the effects of small deviations.
What you might not notice in rigging for example as being too slight a change, washout in particular can be enough in that small size to have an effect.
You'd need a much larger plane, so the deviations could be seen, and preferably planned for during construction. Then fly the pants off the basic airplane to get a feel (most of this stuff is feel anywat) for how it responds. Then make the alterations.
In the end (plane surviving) you'll have an indication as to what deviation creates what response.. on THAT airplane! :D No other plane you build will duplicate all the features: rigging angles, weight disturbances, deviations from "normal" at various points, how those deviations respond to changes in speed..
And also in the end, you'll find that trim can make the airplane fly acceptably, until the configuration gets too weird.
Then configuration alterations and trims will let you fly it more.
The chief thing is to fly, fly, fly.. Get a feel for how different airplanes will and can respond.
No one shape will be ideal for all conditions (Kadets excepted, of course.)
.
PJB's Seriously Aeronautical Stuff:
http://home.earthlink.net/~pjburke1/aindex.html
Sparky Paul:
http://www.angelfire.com/indie/aerostuff

[MikeSt]

Originally posted by Sparky Paul
Your sample plane is really too small to be a good indicator of the effects of small deviations.


Sparky,

I just used those numbers instead of saying "Wingspan X, LeftWing halfpan = .9 X .. .9X" ... which is what I really meant.

Scale the plane up as big as you want (747).

The examples I gave were the "obvious" ones (to me anyway), of how a plane could be out of symmetry. I guess you could also throw in twisted wings / fuselage / stab etc ! :D

E.g. One wing shorter. The longer wing is gonna be heavier, so at slow speed the plane will turn in that direction ??????? .. or will it turn in the opposite direction, since there is more lift on that side ?

Also, I'm guessing that "left wing with less chord", or "left wing less chord" will be similar in effect ????

As I mentioned before, I'd imagine this could all lead to some heavy math.

[MikeSt]

Originally posted by Badger
I would also like to point out that some dimensions are more critical than others. An error of 1% in tail boom length will be far less noticable than a 1% error in airfoil shape. Selig did wind tunnel test on airfoils and found large changes in airfoil drag for changes of a few thousands of an inch in leading edge shape for example.


Iinteresting. Would that few thousandths of an inch be more notiecable as the plane size scales up ?

[Sparky Paul]

i5 May, '83. Israeli F-15 in ACM with an A-4.. collided..
F-15 RTB'd. Looking like this...

[Dick Huang]

Mike Stramba,
Every thing depends on the purpose and performance goal of your model. If your mission is to just bore holes in the sky then I would just follow Sparky Paul's advice and fly,fly,fly. If however you reall want to know the maximum velocity of your model let say and how all the parameters affect it then you need the equations to compute the performance and instrumentation on board to measure the effects. This has been done since the Wright Brothers first flew in 1903!
Yours,
Dick Huang:D

[omega blood]

I heard during desert storm an A-10 pilot got a chunk of a wing (how much I don't remember , but it was a lot) and an turbine shot off. He made it home safe and sound. :)

[Hopperfly22]

I'd say the only reason the F-15 made it back is because it has a 1.5:1 thrust to weight ratio. It's the only jet capable of breaking the sound barrier in vetical flight. Thus with one wing it could pretty much hover back to base.
A-10's can lose half a wing and an engine and still fly back. They're flying tanks by all means including many redundant systems. They had one on display in New Orleans that had taken over 450 rounds of small arms fire and flew back. Most of it .50 cal size stuff.

[Sparky Paul]

Originally posted by Hopperfly22
I'd say the only reason the F-15 made it back is because it has a 1.5:1 thrust to weight ratio. ...
.
Not hardly!
The flight control system is designed with enough redundancy.. seperate power and operating systems for surfaces that permit loss of a significant part of the airplane without losing all control.
The same for the A-10, and even transports like the L-1011 have their flight controls seperated to prevent some failures from crashing the plane.
Extreme over-power didn't (couldn't) do it.

[Lancair]

I used to fly a Cessna 182 G model, on skydiver dropping missions. Its owner had tried outclimbing a CB (thunderstorm) oneday and was lucky to get out alive with aircraft in one piece. BUT end result was a bent wing or two.
The bent section was most easily seen from above , rippling in the skin in parts. In normal ops, ie climbing with a load of skydivers at 75 -80 kts, I had the control wheel over about 30 degrees to keep wings level. Once the meat bombs were away I could speed up to near VNE if I was game, not in this 182 though, 130 was plenty fast enough and at that speed I had the control wheel at about 80 degrees with 95 of travel to keep wings level. So right turns were very slow but turning left was quick to roll in if I relaxed the pressure on the wheel. TOO SCARY, many times I heard a weird noise or two on descent, and once over water I was ready to bail out, but landed it instead. I refused to fly that aircraft again. Funilly enough its still flying, but has been repaired and repainted.

Lancair

[Sparky Paul]

Somewhere in aviation archive land is a series of photos of a Mitsubishi A5M (Claude) that had a midair with a Chinese fighter during the unpleasantness in China prior to WWII.
Lost most of the right wing outboard of the landing gear.
Flown back safely to base.
No AFCS then to help keep the pointy end forward, shear arm power!
A lot of this type of success depends on the control rigging. Seperate system, not one continuous cable from one aileron to the other, for instance.
An AFCS that tries to zero out the rates helps also, probably what saved the F-15 and others.

[Jim Finn]

I have read that the F4U- Corsair had a tendency to drop one wing near stall at landing and it was corrected with a short(8") trangle attached to the right wing just above the landing gear on the leading edge. Not much on a full size plane but if you look at photos of them you can just see it. A pretty small adjustment with a noticable effect.

[Al M]

Many years ago i had an early Lanier. It turned out that one wing was longer than the other by about 1 inch. It flew fine until consecutive loops were attempted. the heading changed 90 degrees through 3 loops.